This project will study 150 female spouse caregivers. Ages 55-80 whose husbands are suffering from Alzheimer's disease or a related disorder. Following extensive psychological and physical evaluations at TL (incl. laboratory testing), caregivers will be randomly assigned to 1 of 3 interventions: anger management class (AMC), depression management class (DMC) or a waiting list (treatment as usual) condition (WL). These particular classes were developed because of past research on the salience of these negative affects for effects for caregivers. Each class will run for 10 weeks, when T2 evaluation will (approx. 3 months for WL). Extensive psychological evaluation will continue but the medical evaluation will focus on physical examination and interview with no laboratory testing. At this time WL subjects may take either class of the cice. Other subjects will be evaluated after 6 months (T3) and again 1 year post (T4) with the same psychological and medical indices as at T2. The major hypotheses are: 1) caregivers who participate in either the AMC or DMC will report significantly less psychological distress (particularly in measures of stress, depression, and anger expression style) following the intervention compared to those in the WL: 2) The mechanism of change will be increased self-efficacy perceptions over time in caregivers who are active participants in either class condition; 3) caregivers who participated in either the AMC or DMC will maintain their gains over time compared to WL participants who did not join a class during the 1 year interval- the mechanism for this will be continued use coping skills learned in the intervention programs: 4) there will be significant positive correlations among indices of psychological stress, anger expression style, and physical health status in particular, measured of cardiovascular disease and hypertension--in this status--in Tl. More specifically, higher psychological stress and an hanger-in anger expression style will be significantly correlated with the presence of cardiovascular disease and hypertension, and with the (continuous) Framingham risk factor score for CVD. If supported this will be very hypothesis generating for future research. Finally, repeated medical evaluations over 1 year will permit study of the relationship of physical and mental health variables and will allow us to examine their interactive predictive power with regard to institutional placement of the care-receivers. Given the increasing number of distressed caregivers in this country, results of this study will have significant distressed caregivers in this country, results of this study will have significant implications for clinical care and for policy-making in the future.